Vercel Adds Azure Provider for DeepSeek V4 Pro and Flash

Vercel

Jun 13, 2026

Vercel added Azure as a provider for DeepSeek V4 Pro and V4 Flash on its AI Gateway. Requests now automatically route through Azure for improved failover, or developers can prioritize Azure using the order parameter in the AI SDK. The integration supports Bring Your Own Key authentication and includes no platform fees or inference markups.

View the full update on vercel.com

Vercel Developers

@vercel_dev2d ago

DeepSeek V4 Pro and Flash now available via Azure on AI Gateway. 𝚖𝚘𝚍𝚎𝚕: '𝚍𝚎𝚎𝚙𝚜𝚎𝚎𝚔/𝚍𝚎𝚎𝚙𝚜𝚎𝚎𝚔-𝚟𝟺-𝚙𝚛𝚘', 𝚙𝚛𝚘𝚟𝚒𝚍𝚎𝚛𝙾𝚙𝚝𝚒𝚘𝚗𝚜: { 𝚐𝚊𝚝𝚎𝚠𝚊𝚢: { 𝚘𝚛𝚍𝚎𝚛: ['𝚊𝚣𝚞𝚛𝚎'] } } https://t.co/3DHjSbHVe6

570

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Keep reading

Vercel AI Gateway Automates Provider Selection Based on Cost and Performance

Vercel added a programmatic sort feature to its AI Gateway that ranks model providers by cost, latency, or throughput at the request level. This update moves the logic of AI economics into the infrastructure layer, allowing applications to automatically route to the cheapest or fastest available provider without code changes.

Vercel AI Gateway Recovers Over 1 Trillion Tokens Monthly with Built-in Redundancy

Guillermo Rauch6d ago

Vercel AI Gateway Recovers Over 1 Trillion Tokens Monthly with Built-in Redundancy

Vercel announced its AI Gateway recovers over 1 trillion tokens monthly by providing redundancy and failover mechanisms. The service offers zero markup over model providers, along with zero-data retention enforcement, observability, usage APIs, and caps. This positions the gateway as a resilient and cost-effective solution for deploying AI applications.

Fireworks AI Expands Azure Foundry Catalog With Frontier Reasoning Models

Fireworks AIMay 15

Fireworks AI Expands Azure Foundry Catalog With Frontier Reasoning Models

Fireworks AI added DeepSeek V4 Pro and Kimi K2.6 to Microsoft Azure AI Foundry while expanding provisioned throughput support to the US Data Zone. The update allows enterprise teams to run high-performance open models with guaranteed throughput and data residency within their existing Azure environment.

vLLM Enables Efficient 1M Token Inference with DeepSeek V4 Day-0 Support

vLLMApr 24

vLLM Enables Efficient 1M Token Inference with DeepSeek V4 Day-0 Support

vLLM has launched immediate support for the DeepSeek V4 Pro and Flash models, which natively handle context windows of up to 1 million tokens. By implementing a new long-context attention mechanism, the update allows developers to serve these massive models with significantly lower memory and compute overhead.